Does dreaming of exam failure affect your real-life chances of success?

Why do we dream? It’s still a scientific mystery. The “Threat Simulation Theory” proposes that we dream as a way to simulate real-life threats and prepare ourselves for dealing with them. “This simulation in an almost-real experiential world would train the brain to perceive dangers and rapidly face them within the safe condition of sleeping,” write the authors of a new paper that’s put the theory to the test.

Isabelle Arnulf and her colleagues reasoned that if dreams help simulate future threats, then students who dream about an important exam might expect to see some kind of advantage over students who don’t.

The researchers contacted thousands of first-year students at the end of the day that they sat a very important exam. Their performance would dictate whether they gained access to medical school. Just over 700 of the students agreed to participate and they completed a questionnaire about their dreams and sleep quality the previous evening, and any dreams they’d had about the exam over the course of the university term.

For the 44.1 per cent of the students who reported having a dream the night before their exam, most said they’d dreamt about the exam, and usually this dream featured some kind of problem such as not understanding questions, getting delayed en route or their pen not working. Looking back over the preceding term, 73.4 per cent of the students said they’d had a dream about the exam, with 87.5 per cent of these dreams featuring some kind of problem.

Next, the researchers gained access to the students’ exam results. The students who dreamt about the exam the night before gained better grades than those students who didn’t, and students who had dreamt about the exam at some point during the term achieved higher grades than students who did not report having any such dream.  The more exam dreams a student reported having during the term, the higher their grade tended to be.

These results appear to provide support for the Threat Simulation Theory of dreaming. Less consistent with the theory is the fact that the precise nature of the exam dreams was not related to exam grades. For example, students who dreamt of problems with the exam didn’t score higher grades than those who dreamt of success. However, it’s notable that the top five scorers on the exam all reported dreaming of exam problems, including being late and running out of time.

“As stated by some students, the immediate benefit of dreaming was the drive to address weaknesses in knowledge after awakening, which certainly provided an advantage,” the researchers said. “Additionally, the contrast between the horrible situations experienced in the dreams (appendicitis, lateness, impossibility of competing) and the the more casual reality (good health, appropriate timing and tools) the next morning may desensitise the students to anxiety, which can be reassuring and beneficial for competition.”

While these results are intriguing, we should interpret then with caution. It’s notable that other factors, besides dream content, had a far stronger relationship with students’ performance, such as past academic achievement and being male. Also, rather than dreams causing superior performance, there many be other factors influencing both dream recall and performance. Anxiety or sleep quality and duration would be obvious candidates, but in fact the researchers found no link between these factors and exam grades. A final issue of course is the unreliability of dream recall, especially considering the students were asked to think back over a whole term.


Arnulf, I., Grosliere, L., Le Corvec, T., Golmard, J., Lascols, O., & Duguet, A. (2014). Will students pass a competitive exam that they failed in their dreams? Consciousness and Cognition, 29, 36-47 DOI: 10.1016/j.concog.2014.06.010

further reading
Does sleeping face-down induce more sexual dreams?
Be careful while you sleep – dreams of jealousy and infidelity spell relationship trouble the next day

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.